a16z maps 2026 crypto priorities: stablecoins, AI, privacy

Andreessen Horowitz’s crypto arm published its annual “big ideas”, mapping priorities that include stablecoin on/offramps, onchain asset origination, AI agent identity, privacy-preserving networks, and advances in zero-knowledge computing.

The firm highlights the scale of stablecoins, estimating $46 trillion in transaction volume last year – more than 20 times PayPal and nearly triple Visa – with transfers clearing in under a second for under a cent. The next phase centers on linking digital dollars to existing rails through regional integrations, QR-based payments, card issuance, and global wallet layers. 

Institutional adoption features prominently. Many banks still rely on mainframe-ledger systems, which make upgrades like real-time payments slow and costly. Stablecoins, tokenized deposits, tokenized treasuries, and onchain bonds are presented as ways to launch new products without rewriting core banking software, effectively accelerating modernization efforts.

On real-world assets, partners argue that tokenizing traditional securities often mirrors legacy structures. General partner Guy Wuollet points to expanded use of perpetual futures to represent assets and to debt originated onchain rather than issued offchain and later tokenized, citing lower servicing and back-office costs. Compliance standardization remains the key hurdle flagged in the report.

The outlook links payments and autonomous software. As AI systems execute tasks on intent, programmable settlement could route value like data packets. a16z’s go-to-market team references agent-to-agent payments for data, GPU time, and API calls, and cites prediction markets that clear in seconds as examples of faster value flows.

Identity for machine actors recurs throughout. Sean Neville, the co-founder of Circle and now CEO of Catena Labs, calls for “Know Your Agent” credentials to bind non-human agents to principals, constraints, and liability. Harvard’s Scott Kominers describes AI-assisted research workflows and points to crypto tools that could attribute and compensate model contributions. Investor Liz Harkavy warns that agents consuming content while bypassing ads and subscriptions put pressure on publishers, urging usage-based nanopayments and attribution standards.

Privacy appears as a competitive focus area. General partner Ali Yahya argues that performance is becoming commoditized while “bridging secrets is hard,” creating stickiness for networks that protect transaction data. The report also backs decentralized, open-protocol messaging designed for a post-quantum world.

Data control and verifiability show up in infrastructure forecasts. Mysten Labs co-founder Adeniyi Abiodun promotes “secrets-as-a-service” – onchain-enforced access rules, client-side encryption, and decentralized key management – to support tokenization and agentic systems in regulated sectors. On security, a16z engineer Daejun Park pushes a shift “from code is law to spec is law,” using formal invariants as runtime guardrails that automatically revert unsafe transactions.

Prediction markets are expected to widen in scope and volume, with decentralized governance and AI-assisted oracles proposed for contested resolutions. Stanford political economist Andy Hall views markets as complementary to polling and points to crypto tools that can help prove survey respondents are human.

Media credibility is another theme. a16z editor Robert Hackett describes “staked media,” where commentators lock tokens, tie forecasts to public markets, and build auditable records to support claims in an era of AI-generated content.

On zero-knowledge computing, Georgetown computer scientist and a16z researcher Justin Thaler projects zk virtual machine provers approaching roughly 10,000x overhead by 2026, with memory footprints small enough for phones and throughput that allows a single GPU to generate proofs of CPU execution in real time. The report argues this could make verifiable cloud computing practical for many CPU workloads without rewriting application code.

The outlook closes with operating guidance. General partner Arianna Simpson cautions that many crypto companies have leaned into trading to find revenue and frames it as a way station rather than an endpoint, urging focus on durable products. The firm notes pending U.S. crypto market-structure legislation as a potential catalyst to align legal and technical architectures, with clearer paths for token launches, decentralization, and transparency.